BACKGROUND: Vision-enabled large language models (VE-LLMs) have the potential to provide flexible and explainable medical image interpretation. However, their real-world performance on clinical data, such as 12-lead electrocardiograms (ECGs), has not...
read more